The New Special Issue "Information-Theoretic Methods in Computational Neuroscience" is Open for Submission!

Guest Editors: Dr. Sarah Marzen, Prof. Dr. John Beggs , Dr. Martina Lamberti and Dr. Jared Salisbury

Submit to the Special Issue: https://www.mdpi.com/journal/entropy/special_issues/5VXL7P10J1

Submission deadline: 30 April 2025

Special Issue Information: Information theory has been an invaluable tool for neuroscience since its conception in the 1940s, with successes ranging from quantifying the rate of information transmission of sensory neurons to the highly influential normative theory of efficient coding to characterizing the interactions within neural populations via maximum entropy models. The present era of experimental neuroscience, marked by increasingly high-dimensional neural and behavioral recordings, poses a particular challenge for information theoretic methods, which typically scale poorly with the dimensionality of the data. At the same time, these rich datasets promise to resolve decades-old questions about the nature of the neural code: information-theoretic methods for understanding the purpose of neural systems using normative theories, such as rate-distortion theory and constrained channel capacity calculations, promise to answer the tough questions of what organisms are trying to do and how they’re doing it.

We welcome original research and reviews that focus on the role of information theory in neuroscience in any way, shape, or form. Examples of topics that may be of interest include:

• The inference and usage of maximum entropy models and stimulus-dependent maximum entropy models;
• The estimation, usage, and interpretation of information-theoretic quantities to benchmark how well neural systems communicate information;
• The development of novel information-theoretic quantities for understanding neural systems;
• Explorations of criticality in neural systems, including optimal information processing capabilities and their connection to criticality and its connection to partial information decomposition and other novel information-theoretic quantities;
• New methods for the estimation of information-theoretic quantities or objectives that pertain to neural systems, particularly those methods that scale to high-dimensional data;
• Information-theoretic normative theories such as rate-distortion theory and its variants or noisy constrained channel coding to understand neural systems, especially in the style of the efficient coding hypothesis.

If there are topics we have missed, they must be especially important! We encourage all authors to submit. If you are wondering if your work fits the scope of the Special Issue, please contact us.